PhD student at TU Delft and ErasmusMC. Specializing in AI and medical imaging.
I am currently a PhD student in a joint program between TU Delft and ErasmusMC, focusing on the development of a 5D growth model to study hip joint growth trajectory using generative and fundamental models. My background includes a Master’s in Biomedical Engineering and significant experience in machine learning, medical imaging, and biomedical engineering.
Here are some of my most recent updates:
Select a publication or project to view detailed information:
This paper introduces a novel method for self-supervised region-aware segmentation of COVID-19 CT images using 3D GAN and contrastive learning.
Publication: Shabani, S., Homayounfar, M., Vardhanabhuti, V., Kohi-moghadam, M. (2022). Computers in Biology and Medicine, 149, 106033.
Paper title: CovidCTNet: an open-source deep Learning Approach to Diagnose Covid-19 Using Small Cohort of CT Images
RQ and solution: The primary objective was to classify COVID CT images (2D or 3D) from CAP (community-acquired pneumonia) and Control images. However, due to the limited number of COVID cases at that time, efforts were directed towards the development of a network capable of robust classification with a restricted sample size. The approach involved utilizing an anomaly detection network on a synthetic CT slice (incorporating infection-like noises) to identify genuine infections within the 3D images, followed by their classification. The pipeline is as follows (see more details in the paper).
Code: https://github.com/mohofar/CovidCtNet
My Contribution: Model development and implementation, writing the technical part of the paper.
Constraint: The simulated infection lacks sufficient generality to encompass all lung infections effectively.
Link to publication: https://www.nature.com/articles/s41746-021-00399-3
This paper explores smartwatch battery usage patterns and provides insights into optimizing battery life based on real-world data.
Publication: Homayounfar, M., Malekijoo, A., Visuri, A., Dobbins, C., Peltonen, E., Pinsky, E., ... & Rawassizadeh, R. (2020). Sensors, 20(13), 3784.
This study employs a multimodal deep learning approach to predict systemic diseases based on oral conditions.
Publication: Zhao, D., Homayounfar, M., Zhen, Z., Wu, M. Z., Yin Yu, S., Yiu, K. H., ... & Koohi-Moghadam, M. (2022). Diagnostics, 12(12), 3192.
This paper introduces FEDZIP, a framework designed to enhance communication efficiency in federated learning through compression techniques.
Publication: Malekijoo, A., Fadaeieslam, M.J., Malekijou, H., Homayounfar, M., Alizadeh-Shabdiz, F., & Rawassizadeh, R. (2021). arXiv preprint arXiv:2102.01593.
This paper discusses a competition focused on recognizing mental arousal levels using shared databases and various algorithms.
Publication: Saidi, M., Rezania, S., Khazaei, E., TaghiBeyglou, B., Hashemi, S.S., Kaveh, R., ... Homayounfar, M., et al. (2019). 27th Iranian Conference on Electrical Engineering (ICEE), IEEE.
This study explores a k-space based motion estimation technique for polar fMRI, leveraging transfer learning to improve accuracy.
Publication: Makhsousi, F., Homayounfar, M., Ghaffarzadeh, S., Nasiraei-Moghaddam, A. (2023). ISMRM 2023 Conference.
You can reach out to me via the following channels: